Abstract and Introduction
Vitamin D deficiency is common in women with polycystic ovary syndrome (PCOS), with the 67–85% of women with PCOS having serum concentrations of 25-hydroxy vitamin D (25OHD) <20 ng/ml. Vitamin D deficiency may exacerbate symptoms of PCOS, with observational studies showing lower 25OHD levels were associated with insulin resistance, ovulatory and menstrual irregularities, lower pregnancy success, hirsutism, hyperandrogenism, obesity and elevated cardiovascular disease risk factors. There is some, but limited, evidence for beneficial effects of vitamin D supplementation on menstrual dysfunction and insulin resistance in women with PCOS. Vitamin D deficiency may play a role in exacerbating PCOS, and there may be a place for vitamin D supplementation in the management of this syndrome, but current evidence is limited and additional randomized controlled trials are required to confirm the potential benefits of vitamin D supplementation in this population.
Polycystic ovary syndrome (PCOS) is the most common endocrine disorder in women of reproductive age, presenting in up to 18% of this population. PCOS is characterized by the presence of polycystic ovaries, menstrual dysfunction, infertility and biochemical (elevated androgens) and clinical (hirsutism and/or acne) hyperandrogenism. PCOS is also associated with an increased incidence of cardiovascular disease (CVD) risk factors, including an increased prevalence of subclinical atherosclerosis, type 2 diabetes, dyslipidaemia and impaired glucose tolerance.[3,4] Obesity and insulin resistance are closely linked to the development of PCOS and its clinical features,[2,5,6] and the current obesity epidemic suggests the prevalence of PCOS will rise in future. This will potentially impact negatively on population growth and cardiovascular morbidity and mortality, and consequently, PCOS has become a major public health concern.
Polycystic Ovary Syndrome and Vitamin D
A number of studies have demonstrated associations between vitamin D levels and various PCOS symptoms, including insulin resistance, infertility and hirsutism.[8–12] Vitamin D is thought to influence the development of PCOS through gene transcription, and hormonal modulation influences insulin metabolism and fertility regulation. Evidence suggests that Vitamin D levels are similar in women with and without PCOS;[8,14] however, there have been reports of lower levels[15–17] and higher levels seen in women with PCOS. Several studies have reported low levels of vitamin D in women with PCOS, with average 25-hydroxy vitamin D (25OHD) levels between 11 and 31 ng/ml,[8–10,14,15,18–23] with the majority having values <20 ng/ml (67–85%).[9–11,14,19,20] Vitamin D deficiency is also common in the general population in many parts of the world, with 10–60% of adults having values lower than 20 ng/ml.[24,25] Vitamin D deficiency disrupts the function of all the systems of the body and increases the risk of chronic disease, including physical diseases such as cancer, cardiovascular, autoimmune and infectious diseases; and psychological disorders such as depression and chronic pain.
Vitamin D3 is obtained from the diet or synthesized endogenously through sunlight-induced photochemical conversion of cholesterol to 7-dehydrocholesterol in the skin and subsequently hydroxylation in the liver and kidney. In the skin, 7-dehydrocholesterol undergoes ultraviolet photolysis to form vitamin D3. Vitamin D3 then undergoes two successive hydroxylations, the first of which takes place in the liver and is catalysed by vitamin D-25 hydroxylase to form 25OHD. The second hydroxylation step is regulated by parathyroid hormone (PTH) and mediated by 25-hydroxyvitamin D3 1a-hydroxylase and occurs predominately in the kidney. This second hydroxylation produces the final active metabolite of vitamin D3, which is 1,25-dihydroxyvitamin D3. 1, 25-dihydroxyvitaminD3 circulates bound to vitamin D-binding protein until it reaches its target tissue where it binds to vitamin D receptors to initiate its effect. Vitamin D receptors occupy some 2776 genomic positions and modulate the expression of 229 genes across a range of tissues, thus providing vitamin D with the potential to impact on a range of physiological processes.
When comparing women with PCOS (n = 85) to control women (n = 115) with similar age (30 year) and body mass index (BMI) (27 kg/m2), Mahmoudi et al. found the women with PCOS had a significantly higher vitamin D level (29·3 ng/ml in PCOS women vs 19·4 ng/ml in control women). On the other hand, Li et al. reported lower vitamin D levels, although not significant, in women with PCOS compared with women without PCOS (11 ng/ml in PCOS group vs 17 ng/ml in control group). However, the ovulatory control group (n = 27) was significantly older (35 year) and had a lower BMI (24 kg/m2) compared with the PCOS group (n = 25, 28 year and 31 kg/m2), which have both been shown to influence vitamin D levels. Recently, Wehr et al. also reported lower levels in women with PCOS (n = 545) compared to the control women (n = 145; 25·7 vs 32·0 ng/ml, respectively), and the PCOS women were significantly younger (27 vs 29 years, respectively).
Thus, it seems that the prevalence of vitamin D deficiency is similar in women with and without PCOS, although there is some inconsistency about whether the levels are similar between women with and without PCOS.
Vitamin D Levels Vary With BMI in PCOS
Many studies have reported inverse associations between body weight (BMI, body fat and waist measurements) and serum 25OHD levels in women with PCOS,[8–11,14,23] with reports of levels 27–56% lower in obese women with PCOS compared with nonobese women with PCOS.[8–10,23] A recent study in women with PCOS also found low 25OHD levels were significantly determined by the degree of adiposity (BMI and total fat mass) and were not directly affected by the development of insulin resistance. It is possible that the high prevalence of vitamin D deficiency in women with PCOS is related to obesity as vitamin D is fat soluble and in obesity, a higher proportion is sequestered in adipose tissue, lowering bioavailability. Alternatively obese subjects may spend less time outdoors exposed to sunlight that can lead to insufficient vitamin D biosynthesis in skin. It is also possible that dietary preferences and vitamin D metabolism may differ between obese and non-obese individuals. However, there has also been one study that reported no difference between 25OHD levels across a range of BMIs (average value approximately 25 ng/ml). At present, there are insufficient data available to identify the basis of the difference in vitamin D status between lean and obese women with PCOS.
Vitamin D Levels, Pathogenesis of PCOS and Insulin Resistance
There is some evidence suggesting that vitamin D deficiency may be involved in the pathogenesis of insulin resistance and metabolic syndrome in PCOS (Figure 1).[11,29] The effects of vitamin D are mediated via both genetic and cellular pathways. Vitamin D regulates gene transcription through nuclear vitamin D receptors (VDR) that are distributed across various tissues, including skeleton, parathyroid glands and the ovaries. The pathogenesis of PCOS has been linked to the effects of VDRs (TaqI, BsmI, FokI, ApaI and Cdx2 polymorphisms) on LH and SHBG levels, testosterone levels, insulin resistance and serum insulin levels.[13,15]Vitamin D deficiency increases PTH production, which is regulated through levels of serum calcium and vitamin D, and increased PTH is independently associated with PCOS, anovulatory infertility and increased testosterone. It has been suggested that the combination of vitamin D deficiency and dietary calcium insufficiency (because serum calcium regulates PTH release) may be largely responsible for the menstrual abnormalities associated with PCOS; However, it has been suggested that vitamin D sufficiency is more important than high calcium intake at maintaining desired values of PTH. However, a recent study found that in women with PCOS, a lower calcium intake was independently associated with higher serum testosterone concentrations, suggesting that a low calcium intake might also contribute to the hormonal dysregulation that occurs in PCOS. VDRs play an important role in oestrogen production in the ovary. Vitamin D regulates oestrogen biosynthesis through direct regulation of the expression of the aromatase gene and by maintaining extracellular calcium homoeostasis. Vitamin D-deficient mice have reduced fertility rates and VDR-null mice demonstrated decreased aromatase activity in the ovary and impaired folliculogenesis.[33–35] In human ovarian tissue, 1,25-dihyroxyvitamin D3 stimulation of oestrogen and progesterone production and lack of effect on testosterone production may be explained by the augmentation of aromatase activity by vitamin D. Aromatase gene expression was decreased in PCOS follicles compared to controls, and they had increased levels of LH but decreased follicular production of progesterone and estradiol by preovulatory follicles, possibly because of the hyperluteinized microenvironment of PCOS follicles. As a result of these effects, vitamin D deficiency may exacerbate symptoms of PCOS.
While the exact mechanism underlying vitamin D and insulin resistance is not known, multiple cellular and molecular mechanisms have been proposed to explain the relationship. The biologically active form of vitamin D, 1,25-dihydroxyvitamin D (1,25OHD) may enhance insulin action by enhancing insulin synthesis and release, increasing insulin receptor expression or suppression of proinflammatory cytokines that are believed to mediate insulin resistance. Vitamin D may also mediate insulin sensitivity by improving calcium status, increasing local production of 25OHD, which leads to transcriptional regulation of specific genes or suppressing serum levels of PTH. Conversely, a recent study that used the gold standard for assessing peripheral insulin sensitivity found that vitamin D deficiency was not related to the presence of insulin resistance, but was related to the presence of obesity. While there is substantial evidence supporting a relationship between vitamin D status and insulin sensitivity, further research is needed to understand the mechanisms.
Vitamin D levels have also been negatively associated with insulin resistance (fasting insulin and HOMA-IR);[8–11] however, in several of these studies, the association disappeared when BMI was controlled for.[8,11] A study by Hahn et al. also found that when grouping the women with PCOS according to 25OHD levels, lower levels of 25OHD were associated with insulin resistance and obesity. It has been suggested that obesity may have a confounding role in the relationship between 25OHD and insulin resistance in women with PCOS. However, one study showed that women with PCOS with severe vitamin D deficiency were more insulin resistant, independently of BMI and WHR, and another showed that while 25OHD levels were lower in obese women with PCOS, in both the obese and non-obese women, 25OHD levels were negatively correlated with BMI and HOMA-IR. Wehr et al. also investigated this relationship using a multivariate regression analysis and found 25OHD levels were a significant and independent predictor for HOMA-IR along with BMI. The aforementioned inverse relationships with markers of adiposity, and with insulin resistance, suggest that low vitamin D status might not only be associated with obesity, but also with insulin resistance in women with PCOS.
Despite the demonstrated associations between PCOS symptoms and vitamin D status, because these are based on cross-sectional studies, causation cannot be established. In order to establish causation, randomized controlled intervention trials need to be undertaken, but there have been very few intervention studies evaluating the effects of vitamin D supplementation in women with PCOS ( Table 1 ). Three small uncontrolled studies have investigated the effect of vitamin D supplementation on insulin resistance in obese women with PCOS. The first uncontrolled study (n = 15) showed that treatment with 1 ug/day alphacalcidol (1-α-hydroxyvitamin D3) for 3 months increased insulin secretion. Serum 25OHD values were low at baseline (15·2 ng/ml) and increased to 28·6 ng/ml. This increase in vitamin D status was positively correlated with the increase in insulin secretion after treatment. Another small uncontrolled intervention trial in obese women with PCOS (n = 11) demonstrated a significant decrease in insulin resistance (HOMA-IR; 4·41–3·67) 3 weeks after a single dose of vitamin D (300 000 IU). The single dose significantly increased 25OHD levels from 16·9 to 37·1 ng/ml, and only two women still had levels lower than 25 ng/ml. A recent pilot study in 46 women with PCOS who received 20 000 IU cholecalciferol weekly for 24 weeks demonstrated an increase in 25OHD levels (28·0 ng/ml to 51·3 ng/ml at 12 weeks and 52·4 ng/ml at 24 weeks). There were significant decreases in fasting and stimulated glucose, but fasting and stimulated insulin and HOMA levels were unchanged. The women in this study were relatively lean without severe insulin resistance, and there was no control group, so the data can only be considered to be preliminary in nature. Vitamin D levels were also high at baseline which may have attenuated the effect of vitamin D supplementation.
These studies suggest lower 25OHD levels are associated with higher insulin resistance. Two small uncontrolled intervention studies have indicated that vitamin D therapy may have a beneficial effect on insulin resistance and insulin secretion in obese women with PCOS; however, another uncontrolled pilot study showed no effect was seen in relatively lean women with PCOS without severe insulin resistance. Further investigation is needed in randomized controlled trials to better understand the effect of vitamin D supplementation in women with PCOS.
Vitamin D and Reproductive Function in Women With PCOS
There is accumulating evidence that vitamin D plays an important role in reproductive function. VDRs have been found in the ovary, endometrium and placenta. Vitamin D deficiency is associated with calcium dysregulation, which contributes to the development of follicular arrest in women with PCOS and results in menstrual and fertility dysfunction. VDR-null mutant mice demonstrated impaired folliculogensis. Two observational studies have investigated vitamin D levels in infertile women, some of whom had PCOS. In one study (n = 67), the 13 women with PCOS had much lower levels of vitamin D than infertile women with normal ovulation, and each unit increase in vitamin D (normalized for BMI, ng/ml per kg/m2) reduced the likelihood of PCOS diagnosis by 96%. In another observational study by Ozkan and colleagues (n = 84) of women undergoing IVF (eight women with PCOS), the patients that achieved pregnancy (n = 26) exhibited significantly higher follicular fluid levels of 25OHD, which was an independent predictor of success of an IVF cycle; adjusting for age, BMI, ethnicity and number of embryos transferred. Each ng/ml increase in follicular fluid 25OHD increased the likelihood for achieving pregnancy by 7%. Those with follicular fluid levels in the lowest to middle tertile were 75% less likely to achieve pregnancy compared with those in the highest tertile. There has also been another study that investigated the effect of 25OHD deficiency of clomiphene-citrate stimulation in women with PCOS. This study found that 25OHD deficiency was associated with lower rates of follicle development and pregnancy after stimulation, independent of lower BMI and older age. Suggest possible role of vitamin D supplementation in infertile PCOS women who undergo ovarian stimulation, and however, it is also possible that there is a relationship between calcium metabolism and reproductive function, rather than PCOS per se.
Studies have also examined the effects of vitamin D supplementation on reproductive function ( Table 1 ). A small uncontrolled intervention study investigated whether vitamin D and calcium dysregulation contribute to the development of follicular arrest, resulting in reproductive and menstrual dysfunction. Thirteen vitamin D-deficient women with PCOS (mean 25OHD value was 11·2 ng/ml) were supplemented with vitamin D combined with calcium which increased 25OHD levels to within the normal range (30–40 ng/ml) within 2–3 months of therapy. This resulted in normalized menstrual cycles within 2 months for seven of the nine women with menstrual dysfunction, two women became pregnant and the other four maintained their normal menstrual cycles. This demonstrated a potential for treatment with vitamin D and calcium to normalize cycles in women with PCOS and low levels of vitamin D.
In a randomized clinical trial investigating the effects of calcium–vitamin D and metformin in regulating the menstrual cycle, 60 infertile women with PCOS were randomized to one of the three treatments consisting of 1000 mg calcium + 400 IU vitamin D per day; 1000 mg calcium + 400 IU vitamin D + 1500 mg/day metformin, or 1500 mg/day metformin. The patients were treated for 3 months and followed up for a further 3 months. The number of dominant follicles (≥14 mm) during the 2–3 months of follow-up was higher in the calcium–vitamin D–metformin group than in either of the other two groups (P = 0·03). However, no significant differences were seen in the rates of pregnancy and menstrual regularity, although improvements in menstrual irregularities were more noticeable in the vitamin D–calcium–metformin group. The authors concluded that metformin and calcium–vitamin D could be effective for the treatment of anovulation and oligomenorrhoea in women with PCOS. However, they did not measure serum 25OHD levels before or after the intervention, so the levels of deficiency and the magnitude of change are unknown. A recent uncontrolled pilot study in 46 women with PCOS also observed improvements in reproductive function, with 50% (23/46) of oligo- or amenorrhoeaic women at baseline reporting improvements in menstrual frequency after 24 weeks of weekly cholecalciferol (20 000 IU), which significantly increased 25OHD levels (28·0–52·4 ng/ml).
This suggests there is a relationship between 25OHD and reproductive function, with low levels of serum 25OHD being associated with ovulatory and menstrual irregularities. Higher vitamin D levels were associated with an increased likelihood for successful pregnancies, and there is evidence for a beneficial effect of vitamin D supplementation on menstrual dysfunction. Further evidence is needed to determine whether vitamin D supplementation is beneficial for pregnancy.
Vitamin D and Hyperandrogenism in Women With PCOS
Observational studies have found relationships between markers of hyperandrogenism and vitamin D status. Hirsute women have been shown to have lower 25OHD levels compared to BMI-matched control women (17 vs 29 ng/ml, respectively), and hirsute women with PCOS have lower 25OHD levels compared with women with PCOS without hirsutism (21·4 vs 26·8 ng/ml, respectively). In women with PCOS, 25OHD levels have been positively associated with SHBG[9,11,14] and negatively associated with the degree of hirsutism,[9,11] free androgen index (FAI),[9,14] total testosterone and dehydroepiandrosterone sulphate. Furthermore, SHBG levels were lower in PCOS women with severely deficient vitamin D levels, but this was no longer significant after adjusting for BMI and WHR. A similar result was found in a study by Wehr et al. where a relationship between SHBG and vitamin D status was no longer significant after controlling for BMI, indicating obesity as the common determinant for both SHBG and 25OHD. Hahn et al. also observed a BMI-dependent increase in FAI and hirsutism score and a decrease in SHBG, again suggesting that the relationships between these variables are related to the presence of obesity rather than to vitamin D status.
This suggests there is a relationship between 25OHD and hyperandrogenism. It has been suggested that the correlations between vitamin D status and hyperandrogenism may be due to the reduction in SHBG that results from obesity.[9,14] Limited studies have looked at the effect of vitamin D supplementation on measures of hyperandrogenism ( Table 1 ), and they have shown no changes in levels of testosterone, SHBG and FAI.[20,22] One small uncontrolled study (n = 13) where women with PCOS were supplemented with vitamin D combined with calcium to increase 25OHD levels to within the normal range (30–40 ng/ml) also reported clinical improvement of acne vulgaris in all three women presenting. There were no other clinical improvements in the other signs of hyperandrogenism measured (hirsutism, acanthosis nigricans and alopecia). Randomized controlled trials are needed to investigate the effect of vitamin D supplementation on hirsutism and androgen levels in women with PCOS.
Vitamin D and CVD Risk Factors in Women With PCOS
There is growing evidence that vitamin D deficiency may adversely affect the cardiovascular system. VDRs are present in vascular smooth muscle[45,46] and endothelium, and large cohort studies have shown that vitamin D deficiency is associated with an increased risk of CVD[48–50] and cardiovascular mortality.[51–53] Studies in women with PCOS have also shown adverse relationships between low vitamin D levels and elevations of CVD risk factors other than insulin resistance, including adverse relationships with total cholesterol, systolic and diastolic blood pressure, glucose, C-reactive protein,[11,14] triglycerides, high-density lipoprotein cholesterol (HDL),[11,14] total cholesterol/HDL and leptin. Furthermore, a large observational study (n = 206) in women with PCOS that investigated the association of 25OHD levels and the metabolic syndrome found that women with PCOS and the metabolic syndrome had lower 25OHD levels than PCOS women without these features (17·3 vs 25·8 ng/ml, respectively; P < 0·05). The small uncontrolled intervention studies previously mentioned that investigated insulin resistance also assessed the impact of vitamin D supplementation on other CVD risk factors ( Table 1 ). There were reported improvements in serum triglycerides and HDL, with no changes in BMI and significant decreases in fasting and stimulated glucose, triglycerides and hip circumference, but total cholesterol and low-density lipoprotein cholesterol were increased, and BMI, HDL, blood pressure and waist circumference were unchanged. Another study also found that atorvastatin (20 mg/day for 3 months) increased 25OHD levels in women with PCOS (18·4–24·4 ng/ml) compared to no change in the placebo group, suggesting there may be a different beneficial effect of statins on vitamin D levels in women with PCOS. The increase in 25OHD levels was significantly correlated with the reduction in high-sensitivity C-reactive protein, but not with total cholesterol, triglycerides, HOMA-IR or FAI.
Vitamin D and Psychological Well-being in Women With PCOS
The clinical features and potential health implications associated with PCOS, including menstrual dysfunction, difficulties in conceiving, changes in physical appearance (excessive hair growth, obesity and acne) can promote psychological morbidity reflected by loss of self-esteem, emotional stress and poor body image, all with negative impacts on HRQOL[55–59] and an increased likelihood of experiencing depressive symptoms.[56,59,60] In addition to the psychological responses to PCOS symptoms, there may be a more fundamental link between depression and vitamin D deficiency that is mediated by hormones and neurotransmitters. A review in 2008 found four studies showing an association between low 25OHD levels and higher incidences of mood disorders (premenstrual syndrome, seasonal affective disorder, non-specified mood disorder and major depressive disorder). It is possible that these results are confounded by the fact that simply getting out into the sun can have a positive effect on mood and help increase 25OHD levels.
Intervention studies in vitamin D-deficient patients have also shown a clinically significant reduction in depressive symptoms of up to 50% when supplemented with vitamin D.[62,63] However, to date, no studies have investigated whether there is a link between vitamin D and psychological well-being or any effects of vitamin D supplementation in women with PCOS. This is a potential area that requires further investigation.
Clin Endocrinol. 2012;77(3):343-350. © 2012 Blackwell Publishing